It is increasingly recognised that chronic fatigue syndrome (CFS) is heterogeneous. A significant proportion of patients fulfilling operative criteria for a diagnosis of CFS will also fulfill criteria for a psychiatric disorder, such as depression or somatisation. Failure to recognise this heterogeneity prejudices attempts to understand CFS in cross sectional studies. In this issue (pp 302–307) Fulcheret al report a study of muscle strength, aerobic exercise capacity, and functional incapacity in a group of patients with CFS without concurrent psychiatric disorder, compared with patients with major depression and a group of normal but sedentary subjects.1 In an incremental treadmill exercise test, patients with CFS and depressed patients had lower peak oxygen consumption rates, maximal heart …